End of training

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 15.0,
-    "eval_loss": 0.3827837407588959,
-    "eval_runtime": 112.0874,
+    "eval_loss": 0.3709133565425873,
+    "eval_runtime": 115.3267,
     "eval_samples": 1647,
-    "eval_samples_per_second": 14.694,
-    "eval_steps_per_second": 1.838,
-    "eval_wer": 0.34950464712491064,
-    "train_loss": 1.0725414783582774,
-    "train_runtime": 6951.8597,
+    "eval_samples_per_second": 14.281,
+    "eval_steps_per_second": 1.786,
+    "eval_wer": 0.3428658972525789,
+    "train_loss": 1.0641350445761957,
+    "train_runtime": 6976.4654,
     "train_samples": 3478,
-    "train_samples_per_second": 7.504,
-    "train_steps_per_second": 0.235
+    "train_samples_per_second": 7.478,
+    "train_steps_per_second": 0.234
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 15.0,
-    "eval_loss": 0.3827837407588959,
-    "eval_runtime": 112.0874,
+    "eval_loss": 0.3709133565425873,
+    "eval_runtime": 115.3267,
     "eval_samples": 1647,
-    "eval_samples_per_second": 14.694,
-    "eval_steps_per_second": 1.838,
-    "eval_wer": 0.34950464712491064
+    "eval_samples_per_second": 14.281,
+    "eval_steps_per_second": 1.786,
+    "eval_wer": 0.3428658972525789
 }

runs/Jul26_07-42-20_8b1495053dbb/events.out.tfevents.1690365097.8b1495053dbb.1511.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8db96e9cbabca7391a632d90e73681ba7f2fc69d97c0b5db0b7d74e48eeff4a
+size 406

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 15.0,
-    "train_loss": 1.0725414783582774,
-    "train_runtime": 6951.8597,
+    "train_loss": 1.0641350445761957,
+    "train_runtime": 6976.4654,
     "train_samples": 3478,
-    "train_samples_per_second": 7.504,
-    "train_steps_per_second": 0.235
+    "train_samples_per_second": 7.478,
+    "train_steps_per_second": 0.234
 }

trainer_state.json

@@ -9,174 +9,174 @@
   "log_history": [
     {
       "epoch": 0.92,
-      "eval_loss": 3.5826058387756348,
-      "eval_runtime": 112.5008,
-      "eval_samples_per_second": 14.64,
-      "eval_steps_per_second": 1.831,
+      "eval_loss": 3.5988094806671143,
+      "eval_runtime": 113.2606,
+      "eval_samples_per_second": 14.542,
+      "eval_steps_per_second": 1.819,
       "eval_wer": 1.0,
       "step": 100
     },
     {
       "epoch": 1.83,
-      "eval_loss": 3.0217819213867188,
-      "eval_runtime": 113.1278,
-      "eval_samples_per_second": 14.559,
-      "eval_steps_per_second": 1.821,
+      "eval_loss": 3.008251905441284,
+      "eval_runtime": 111.5132,
+      "eval_samples_per_second": 14.77,
+      "eval_steps_per_second": 1.847,
       "eval_wer": 0.9998978653865795,
       "step": 200
     },
     {
       "epoch": 2.75,
-      "eval_loss": 0.8984940052032471,
-      "eval_runtime": 111.3066,
-      "eval_samples_per_second": 14.797,
-      "eval_steps_per_second": 1.851,
-      "eval_wer": 0.8035951383924012,
+      "eval_loss": 0.8641791939735413,
+      "eval_runtime": 113.049,
+      "eval_samples_per_second": 14.569,
+      "eval_steps_per_second": 1.822,
+      "eval_wer": 0.757940966193443,
       "step": 300
     },
     {
       "epoch": 3.67,
-      "eval_loss": 0.5992483496665955,
-      "eval_runtime": 112.2673,
-      "eval_samples_per_second": 14.67,
-      "eval_steps_per_second": 1.835,
-      "eval_wer": 0.619650699622102,
+      "eval_loss": 0.5713350176811218,
+      "eval_runtime": 113.001,
+      "eval_samples_per_second": 14.575,
+      "eval_steps_per_second": 1.823,
+      "eval_wer": 0.6202635073026248,
       "step": 400
     },
     {
       "epoch": 4.59,
       "learning_rate": 0.00029939999999999996,
-      "loss": 3.1629,
+      "loss": 3.14,
       "step": 500
     },
     {
       "epoch": 4.59,
-      "eval_loss": 0.49684980511665344,
-      "eval_runtime": 112.6664,
-      "eval_samples_per_second": 14.618,
-      "eval_steps_per_second": 1.828,
-      "eval_wer": 0.5339597589623123,
+      "eval_loss": 0.47950074076652527,
+      "eval_runtime": 113.7593,
+      "eval_samples_per_second": 14.478,
+      "eval_steps_per_second": 1.811,
+      "eval_wer": 0.5337554897354714,
       "step": 500
     },
     {
       "epoch": 5.5,
-      "eval_loss": 0.4645964503288269,
-      "eval_runtime": 111.3096,
-      "eval_samples_per_second": 14.797,
-      "eval_steps_per_second": 1.851,
-      "eval_wer": 0.5045449902972117,
+      "eval_loss": 0.4440889060497284,
+      "eval_runtime": 114.5305,
+      "eval_samples_per_second": 14.38,
+      "eval_steps_per_second": 1.799,
+      "eval_wer": 0.49116535593912775,
       "step": 600
     },
     {
       "epoch": 6.42,
-      "eval_loss": 0.43159034848213196,
-      "eval_runtime": 113.1768,
-      "eval_samples_per_second": 14.552,
-      "eval_steps_per_second": 1.82,
-      "eval_wer": 0.44254927995097537,
+      "eval_loss": 0.4241042733192444,
+      "eval_runtime": 115.6017,
+      "eval_samples_per_second": 14.247,
+      "eval_steps_per_second": 1.782,
+      "eval_wer": 0.4521499336125013,
       "step": 700
     },
     {
       "epoch": 7.34,
-      "eval_loss": 0.44998037815093994,
-      "eval_runtime": 111.8678,
-      "eval_samples_per_second": 14.723,
-      "eval_steps_per_second": 1.841,
-      "eval_wer": 0.4734960678173833,
+      "eval_loss": 0.4326329827308655,
+      "eval_runtime": 113.9718,
+      "eval_samples_per_second": 14.451,
+      "eval_steps_per_second": 1.807,
+      "eval_wer": 0.4611377795935042,
       "step": 800
     },
     {
       "epoch": 8.26,
-      "eval_loss": 0.41139060258865356,
-      "eval_runtime": 113.568,
-      "eval_samples_per_second": 14.502,
-      "eval_steps_per_second": 1.814,
-      "eval_wer": 0.41231743437851087,
+      "eval_loss": 0.39125096797943115,
+      "eval_runtime": 115.5103,
+      "eval_samples_per_second": 14.258,
+      "eval_steps_per_second": 1.783,
+      "eval_wer": 0.42120314574609335,
       "step": 900
     },
     {
       "epoch": 9.17,
       "learning_rate": 0.00016810572687224668,
-      "loss": 0.2226,
+      "loss": 0.2183,
       "step": 1000
     },
     {
       "epoch": 9.17,
-      "eval_loss": 0.41622716188430786,
-      "eval_runtime": 112.362,
-      "eval_samples_per_second": 14.658,
-      "eval_steps_per_second": 1.833,
-      "eval_wer": 0.4018997038096211,
+      "eval_loss": 0.40361154079437256,
+      "eval_runtime": 114.1472,
+      "eval_samples_per_second": 14.429,
+      "eval_steps_per_second": 1.805,
+      "eval_wer": 0.39730364620569913,
       "step": 1000
     },
     {
       "epoch": 10.09,
-      "eval_loss": 0.3999302089214325,
-      "eval_runtime": 113.8252,
-      "eval_samples_per_second": 14.47,
-      "eval_steps_per_second": 1.81,
-      "eval_wer": 0.38239199264630785,
+      "eval_loss": 0.40348684787750244,
+      "eval_runtime": 113.8647,
+      "eval_samples_per_second": 14.465,
+      "eval_steps_per_second": 1.809,
+      "eval_wer": 0.3958737616178123,
       "step": 1100
     },
     {
       "epoch": 11.01,
-      "eval_loss": 0.404796838760376,
-      "eval_runtime": 112.9341,
-      "eval_samples_per_second": 14.584,
-      "eval_steps_per_second": 1.824,
-      "eval_wer": 0.38423041568787664,
+      "eval_loss": 0.38071152567863464,
+      "eval_runtime": 112.8863,
+      "eval_samples_per_second": 14.59,
+      "eval_steps_per_second": 1.825,
+      "eval_wer": 0.3790215504034317,
       "step": 1200
     },
     {
       "epoch": 11.93,
-      "eval_loss": 0.3789042532444,
-      "eval_runtime": 112.9251,
-      "eval_samples_per_second": 14.585,
-      "eval_steps_per_second": 1.824,
-      "eval_wer": 0.3602287815340619,
+      "eval_loss": 0.3750300109386444,
+      "eval_runtime": 113.6392,
+      "eval_samples_per_second": 14.493,
+      "eval_steps_per_second": 1.813,
+      "eval_wer": 0.36502910836482483,
       "step": 1300
     },
     {
       "epoch": 12.84,
-      "eval_loss": 0.4024040102958679,
-      "eval_runtime": 114.3571,
-      "eval_samples_per_second": 14.402,
-      "eval_steps_per_second": 1.801,
-      "eval_wer": 0.35359003166173014,
+      "eval_loss": 0.38215482234954834,
+      "eval_runtime": 114.233,
+      "eval_samples_per_second": 14.418,
+      "eval_steps_per_second": 1.803,
+      "eval_wer": 0.35726687774486776,
       "step": 1400
     },
     {
       "epoch": 13.76,
       "learning_rate": 3.594713656387665e-05,
-      "loss": 0.1015,
+      "loss": 0.1011,
       "step": 1500
     },
     {
       "epoch": 13.76,
-      "eval_loss": 0.38988158106803894,
-      "eval_runtime": 113.4392,
-      "eval_samples_per_second": 14.519,
-      "eval_steps_per_second": 1.816,
-      "eval_wer": 0.35747114697170873,
+      "eval_loss": 0.37470725178718567,
+      "eval_runtime": 113.9728,
+      "eval_samples_per_second": 14.451,
+      "eval_steps_per_second": 1.807,
+      "eval_wer": 0.35103666632621794,
       "step": 1500
     },
     {
       "epoch": 14.68,
-      "eval_loss": 0.3801647126674652,
-      "eval_runtime": 112.3523,
-      "eval_samples_per_second": 14.659,
-      "eval_steps_per_second": 1.834,
-      "eval_wer": 0.3489939740578082,
+      "eval_loss": 0.37135428190231323,
+      "eval_runtime": 113.4186,
+      "eval_samples_per_second": 14.521,
+      "eval_steps_per_second": 1.816,
+      "eval_wer": 0.3454192625880911,
       "step": 1600
     },
     {
       "epoch": 15.0,
       "step": 1635,
       "total_flos": 6.468063251673315e+18,
-      "train_loss": 1.0725414783582774,
-      "train_runtime": 6951.8597,
-      "train_samples_per_second": 7.504,
-      "train_steps_per_second": 0.235
+      "train_loss": 1.0641350445761957,
+      "train_runtime": 6976.4654,
+      "train_samples_per_second": 7.478,
+      "train_steps_per_second": 0.234
     }
   ],
   "max_steps": 1635,